Childhood adversities have been reported to be more common among individuals at ultra-high risk (UHR) for psychosis. This paper systematically reviewed and meta-analysed (i) the severity and prevalence of childhood adversities (childhood trauma exposure, bullying victimisation and parental separation or loss) among the UHR, and (ii) the association between adversities and transition to psychosis (TTP). PsycINFO, PubMed and Embase databases were searched for studies reporting childhood adversities among UHR individuals. Only published articles were included. Risk of bias was assessed using Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline (von Elm et al., 2007) and the tool developed by Hoy et al. (2012). Seventeen case–control, cross-sectional and longitudinal studies were included. UHR individuals experienced significantly more severe trauma than controls, regardless of trauma subtype. UHR were 5.5, 2.5 and 3.1 times as likely to report emotional abuse, physical abuse and bullying victimisation, respectively. There was no association with parental separation. However, childhood trauma was not significantly associated with TTP (follow-up periods: 6 months to 15 years), suggesting that trauma alone may not be a sufficient risk factor. Sexual abuse was associated with TTP but this may have been driven by a single large study. Potential confounders and low rates of TTP among UHR are limitations of this review. This is the first meta-analysis that quantitatively summarises the associations between childhood adversities and TTP among UHR, and between specific abuse subtypes and TTP. Specific recommendations have been made to increase the quality of future research. PROSPERO registration no. CRD42017054884.